Wednesday, August 5, 2009 - 10:10 AM

COS 64-7: The impacts of ecological restoration treatments on wildlife populations in ponderosa pine forests: A meta analysis

Elizabeth L. Kalies and Wally Covington. Northern Arizona University

Background/Question/Methods

After a century of fire suppression, grazing, and logging, ponderosa pine forests in the western United States have undergone a radical departure from previous “natural conditions” and are susceptible to stand-replacing wildfire.  There is now an emphasis on implementing restoration treatments in ponderosa pine forests, whereby fuel loads are reduced through both silvicultural treatments and controlled burns.  These treatments should, in theory, restore the native, diverse assemblage of animal species.  However, both thinning and burning treatments are being implemented across thousands of acres in the ponderosa pine forests of the western United States, with limited understanding of the implications to wildlife.  Individual species have been studied, but no review exists that quantitatively analyzes the existing literature across taxa.  The objective of this meta analysis is to evaluate the impacts of ecological restoration treatments on wildlife populations in ponderosa pine forests in the western United States.  Our approach uses meta analysis and model selection to determine the characteristics of treatments with the greatest impact on wildlife densities, as well as species that are least and most sensitive to habitat manipulation. 

Results/Conclusions

We show that while some guilds respond positively to treatments (including aerial- and ground-foraging birds), others respond negatively (tree squirrels).  We also identify species for which there is a paucity of field experimentation and data in the literature, or too much variation in outcomes to reach a conclusion.  While the type of treatment has little effect on animal densities, the time since treatment does have an effect.  Finally, we show that the method of measuring wildlife density does not affect the variability between studies, but before-after studies have more variability than treatment-control methods.  Thus, we provide direction for both methodological approaches to wildlife population research, and species in need of further study.